Alcoholics and drug abusers cite psychological stress and negative mood as a common reason for continued drug use and relapse. However, there is little understanding of the brain mechanisms by which exposure to stress and negative mood results in continued alcohol use and an increased risk of relapse. Using animal models of relapse, several studies have shown that acute stress exposure reinstates alcohol-seeking behavior in alcohol dependent laboratory animals. This research also suggests that stress-induced reinstatement of drug seeking is associated with activation of brain stress circuits, namely corticotrophin releasing factor (CRF)-hypothalamicpituitary- adrenal (HPA) axis and central noradrenergic/sympathoadrenomedullary systems. Our previous work with cocaine dependent individuals and cocaine alcoholics has shown that imagery of personal stressors and of alcohol/drug cue situations as compared to neutral-relaxing imagery reliably increases drug craving and physiological arousal. Such increases in drug craving are accompanied by activation of HPA measures such as adrenocorticotrophic hormone (ACTH), cortisol and prolactin, and increases in plasma catecholamines along with sympathetic arousal. Most recently, stress-induced increases in cortisol and plasma catecholamines were found to significantly differentiate between cocaine relapsers and non-relapsers and also predict time to cocaine relapse after inpatient cocaine treatment. However, there is no previous research examining whether the above brain stress circuits are activated when alcoholics are exposed to personal stressors and to alcohol cues. Whether the response of the above brain stress circuits to stress and to alcohol cues is associated with alcohol relapse in humans has also not been examined thus far. Thus, in a sample of 50 healthy controls and 100 treatment-engaged alcoholics in early recovery (4-8 weeks abstinent), we propose a 5-year study to address the following specific aims: (1) To examine whether imagery of personal stress situations and of alcohol cue situations as compared to neutral situations results in significant increases in alcohol craving, changes in emotion state, HPA activation, plasma catecholamines and physiological arousal; (2) To assess if alcoholics and healthy controls differ in their reactivity to stress and to alcohol cue imagery as compared to neutral imagery; (3) To examine whether brain stress circuit response to stress and to alcohol cues is associated with alcohol relapse; and (4) To examine whether demographic and individual difference variables such as race, gender, age, psychiatric co-morbidity, frontal executive functioning, stressful life experiences, drinking history and alcohol dependence severity are significantly associated with stress and alcohol cue reactivity and alcohol relapse? The findings from the proposed study will provide a greater understanding of the mechanisms by which stress and alcohol cues increase alcohol craving and relapse risk, and will have significant implications for the development of new pharmacological and behavioral interventions for alcoholism.